|Publication number||US7538281 B2|
|Application number||US 11/271,685|
|Publication date||May 26, 2009|
|Filing date||Nov 10, 2005|
|Priority date||May 13, 2003|
|Also published as||US7009118, US20040226755, US20060064264, WO2004102134A2, WO2004102134A3|
|Publication number||11271685, 271685, US 7538281 B2, US 7538281B2, US-B2-7538281, US7538281 B2, US7538281B2|
|Inventors||James R. Pottebaum, Amar Inalsingh|
|Original Assignee||The Ais Group, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (15), Referenced by (11), Classifications (10), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is a Continuation of application Ser. No. 10/437,337, filed May 13, 2003 now U.S. Pat. No. 7,009,118.
The present invention relates to weighing loads placed on trucks and other vehicles, and, in particular, to vehicle load weighing systems that are sufficiently accurate to enable haulers to accurately bill customers for incremental loads that are picked up over the course of a day.
Load cells have been used in vehicle load weighing systems. Many designs of load cells have been used for detection of overload conditions in trucks for safety reasons. These systems were not used for customer billing because these systems did not have sufficient accuracy or precision. It was only necessary that the accuracy be great enough to detect an unsafe load level.
Known load cells utilize strain gauges which may include Wheatstone bridge arrays of fine wires which change resistance when under strain. Examples of related prior art are seen in U.S. Pat. No. 3,661,220 to Harris; U.S. Pat. No. 3,734,216 to Nordstrom et. al; U.S. Pat. No. 3,857,452 to Hartman; U.S. Pat. No. 4,020,911 to English et al.; U.S. Pat. No. 4,364,279 to Stern et al.; and U.S. Pat. No. 4,666,003 to Reichow.
Typically, prior art weighing systems used load cells that provided analog outputs to either central control boxes or to junction boxes where the analog signals representing weight would be converted to digital signals. Examples of these systems are U.S. Pat. No. 4,020,911 to English et al.; U.S. Pat. No. 4,456,084 to Miller. In a junction box system, the outputs of a number of load cells would be connected to a junction box which contains analog to digital converters, memory and a processor. The junction box would either sum the analog voltages representing weight and then convert the summed voltage representing total weight to a digital signal, or, alternately, convert the individual analog voltages to digital signals, and then sum the digital signals to obtain an overall weight. The digital signal indicating weight would then be transmitted to a system controller or display. One problem with junction box systems is that the analog signal that represents weight is subject to loss and noise between the load cell and the junction box. This limits the accuracy that can be achieved with these systems.
Some prior art load cells contained non-volatile memories to hold calibration data for the load cell. A microprocessor in the load cell would use the calibration data to adjust the gain and offset of the load cell, as in U.S. Pat. No. 5,710,716 to Hurst et al. The method of U.S. Pat. No. 5,814,771 to Oakes et al. is to calibrate a junction box on each trailer with the load cells on that trailer and store the calibration constants in a non-volatile memory in the junction box. The method of the Hurst et al. '716 patent is to issue a command from the system controller 40 to the load cell to cause the load cell to enter a zero setting mode and a calibration setting mode. In the zero setting mode switches 32 apply a null voltage to amplifier 30. Analog to digital converter 26 reads the output voltage of amplifier 30 and micro-controller 20 determines an appropriate correction voltage to apply using a digital to analog converter connected to the non-inverting input of amplifier 30. A second zero is performed to zero the strain gauge when the vehicle is unloaded. A gain setting mode is entered into under control of the system controller when the truck has a known load, usually when it is on a scale. The gain setting mode determines a constant which is multiplied by the reading of the analog to digital converter to determine the load weight. No allowance for temperature changes is made. The limited error correction that these systems are capable of performing does not enable these systems to be of sufficient accuracy for customer billing.
U.S. Pat. No. 5,478,974 to O'Dea discloses an on-board vehicle weighing system that includes a “Tare” mode for display. The system, however, uses the pressure of air in the air suspension for the trailer and does not use high precision load cells. As such, the O'Dea system cannot achieve the accuracy of the present invention.
One prior art system, U.S. Pat. No. 5,789,714 to Doerksen, requires that the load be raised by hydraulic pistons in order to weigh it, and indicates that added time for the measurement to settle is required due to vibration caused by the lifting.
These prior art systems suffer from lack of accuracy and lack of resolution needed to maintain accurate customer billings. These systems are also susceptible to mechanical and electrical noise causing weight measurements to be degraded. Measurement drift due to temperature changes and creep of the strain gauge over time also degrades results.
Prior art on-board load weighing systems have not been certified or certifiable by weights and measures agencies because they were not sufficiently accurate for customer billing.
One object of the invention is to provide a load cell which has greater resolution than prior load cells.
Another object of the invention is to improve the usable resolution of a load cell by using sophisticated temperature compensation and linearity compensation to correct for strain gauge errors.
Another object of the invention is to improve the accuracy of a load cell by using sophisticated formulas or piecewise linearization for temperature compensation and linearity compensation to better convert raw strain gauge readings into weight readings.
Another object of the invention is to reduce errors due to mechanical and electrical noise on measurements by using adaptive, dynamic digital filtering of the measured strain data.
Another object of the invention is to improve the accuracy of the truck load weighing system so that it can be used for accurate customer billing on loads that are picked up.
Another object of the invention is to reduce maintenance cost of the system by using built-in fault isolation means and built-in fault detection. Position locators will be used so that failed load cells will be identified by position on the truck, thus lowering the mean time to repair any failed load cells.
Another object of the invention is to make the truck load weighing system tamper resistant by storing calibration history and data in non-volatile RAM in the load cells and the controller.
Another object of the invention is to reduce cost and improve reliability of the truck load weighing system by minimizing wiring connections using a single communication bus that connects all system components to the system controller.
Another object of the invention is to simplify maintenance by storing calibration, configuration and manufacturing history information in the load cell.
Another object of the invention is to allow factory monitoring of system and load cell performance through the cab display, direct connection or wireless LAN, so that failures can be detected early and performance monitored to detect imminent failures.
Another object of the invention is to perform incremental weighing. Because of the increased accuracy and resolution of the load cells, sufficient accuracy is available to weigh individual load pick-ups and calculate the total truck weight using the totals of the individual pick-ups.
Another object of the invention is to use multiple ranges for weighing loads. Because of the added resolution available due to the use of high resolution analog-to-digital converters and the use of temperature and linearity compensation, the present system can measure pick-ups in varying resolution based on the size of the pick-up. A resolution of 10 lbs can be used for a pick-up of 50,000 lbs; a resolution of 2 lbs. for a pick-up of 10,000 lbs or less; and a resolution of 1 lb or less for a pick-up of 1,000 lbs.
Another object of the invention is to compensate for the angle of the vehicle due to one or more wheels being at a different elevation than the other wheels to reduce the measured error due to tilt.
These objects of the invention are accomplished by a system that comprises novel and innovative load cells combined with an innovative system controller implementing a cab display wherein the load weighing system has three or more load cells.
In a preferred embodiment, the load cells avoid operational problems with prior art load cells by using a Wheatstone bridge strain gauge connected to a high resolution (21 bit or better resolution) analog-to-digital converter through low offset amplifiers. The readings of the high resolution analog-to-digital converter are filtered through an adaptive digital filter implemented in software running on a microprocessor located in the load cell to minimize the deleterious effects of mechanical and electrical noise on the measured weight. The filtered raw weight data is then compensated for thermal effects on data zero using third order compensation formulae. This zero compensated data is then compensated for thermal effects on gain using another third order equation.
Filtered and thermally compensated data is sent from the individual load cells that are mounted between the truck chassis and the container to a central cab display and controller which performs the final processing of the data. The load cells are all connected to the cab display by a multi-drop network, such as a controller area network (CAN). This network allows for future expansion of the load weight measuring system and also allows for the connection of other necessary sensors, such as a tilt sensor. A tilt sensor is used to provide a measurement of the angle that the truck has with respect to level so that the cab controller can compensate the load cell measurements for a truck parked on sloping ground.
In the present invention a plurality of load cells are used between the vehicle container and its supporting chassis. The number of load cells used varies from as few as three which would define a triangular base to six as seen in the drawings herein, or to a greater number as required. This invention is applicable to many types of vehicles including but not limited to garbage trucks, semi tractor trailer trucks, flatbeds and helicopters. As used with trucks a typical load cell of the present invention is secured beneath a bottom surface of the truck container and upon a mounting surface of a beam of the truck chassis.
In a preferred embodiment of the new load cell a new dual shear beam configuration is established. The load cell's elongated block is situated such that it is supported from below at its opposite ends and a load is applied to the top of the block midway between the ends, thus establishing a system of a central downward force and upward reaction forces at the ends. Between the first upward force and the central downward force is a first strain gauge location, and between the central downward force and the second upward force is a second strain gauge location. At each of these locations the block has front and rear faces extending generally vertically and a Wheatstone bridge arrangement of strain gauges is fixed to each of said front and rear faces. Thus, on the front face are two Wheatstone bridge arrangements on the two strain gauge areas between the downward force and upward forces respectively. On the rear face are similar strain gauge areas directly behind and opposite those on the front face.
In a preferred embodiment in each strain gauge area, the load cell block has opposing recesses in the front and rear faces, such that the back walls of the two opposing recesses are spaced apart and define between them a web or internal vertical wall. A Wheatstone bridge arrangement of four strain gauges is adhered to each of the front and rear faces of this wall. Since the block has two such walls, there are a total of sixteen strain gauges.
With this arrangement of arrays of strain gauges on the spaced apart front and rear faces of the two internal webs, axial, transverse and twisting forces that might be applied to the block and to the strain gauges thereon due to thermal distortion of the truck container and of the truck chassis or due to tilting and any other loading conditions, will be fully or substantially cancelled. The result will be that only the shear load is experienced and measured by the load cells, leading to significantly more accurate results.
Another embodiment of the present invention constitutes a method of measuring a weight comprising a housing, a plurality of Wheatstone bridges each comprising four strain gauges, each Wheatstone bridge having input terminals and output terminals, the inputs of said plurality of Wheatstone bridges being connected to a reference voltage and a ground plane and the output analog signals being indicative of strain applied to said load cell, a plurality of amplifiers receiving and amplifying output from said plurality of Wheatstone bridges, an analog-to-digital converter receiving and converting output of said amplifiers into a digital format indicative of said strain, a programmable read only memory, a network interface, a microprocessor receiving outputs from said analog-to-digital converter, and software programs resident in the programmable read only memory for adaptively filtering digital data, and for executing at least one of the routines of: adjusting digital data for factory setting, adjusting for coarse gain correction and compensating for non-linearity. The microprocessor executes the software programs resident in the programmable read only memory which adaptively filters the digital data, and executes at least one of the following routines: adjusts the digital data for the factory zero, performs said coarse gain adjustment, performs non-linearity compensation on the digital data, and averages said filtered and compensated strain data from the four strain gauges of each of the Wheatstone bridges, and the microprocessor outputs said filtered and compensated digital data using said network interface.
Another embodiment of the present invention is for a load weighing system for weighing individual loads put onto a container carried by the chassis of vehicle comprising: at least three high precision load cells, each including at least one Wheatstone bridge strain gauge, system controller software, network wiring and a system controller comprising: processing means, programmable read only memory, network interface circuit means and user interface means. The high precision load cells are mounted between the vehicle chassis and the vehicle container to provide data from the Wheatstone bridge strain gauges indicative of weight of the container and load thereon, the high precision load cells and the system controller are connected electronically by the network wiring, the system controller receives commands from a user on a user interface, the system controller issues commands using the network interface for directing said load cells to report weight, and the load cells transmit filtered and compensated digital strain data to the system controller which calculates and displays weight.
Another embodiment of the present invention is for a method of measuring a load weight comprising: (a) initiation by an operator of the load weight measurement before a load is added, receiving data reflecting weight on a plurality of load cells from said plurality of load cells, receiving data reflecting two axes of tilt of the truck container, calculating the initial truck container weight from the plurality of load cell data and the two axes of truck container tilt; and (b) initiation by an operator of a load weight measurement after the load is added, receiving data reflecting weight on a plurality of load cells from said plurality of load cells, receiving data reflecting two axes of tilt of the truck container, calculating the initial truck container weight from the plurality of load cell data and the two axes of truck container tilt, and displaying the load weight on an operator display.
Another embodiment of the present invention is for a method of measuring a load weight comprising: (a) initiation of the load weight measurement before a load is added, receiving data reflecting weight on a plurality of load cells from said plurality of load cells, receiving data reflecting two axes of tilt of the truck container, calculating the initial truck container weight from the plurality of load cell data and the two axes of truck container tilt; and (b) initiation of a load weight measurement after the load is added, receiving data reflecting weight on a plurality of load cells from said plurality of load cells, receiving data reflecting two axes of tilt of the truck container, calculating the initial truck container weight from the plurality of load cell data and the two axes of truck container tilt, and displaying the load weight on an operator display.
Another embodiment of the present invention is for a vehicle load weight measuring system comprising a plurality of load cells with serial communications capability to transmit measurements of weight, and a controller to receive weight measurements from the plurality of load cells. The load cells transmit weight measurements to the controller and the controller calculates the weight.
A further embodiment of the present invention is a load cell operable with a vehicle having a chassis part and a container part carried by said chassis part, where said load cell is supported by said chassis part and supports said container part for measuring the weight of said container part and any load therein. This load cell comprises:
A still further embodiment of the present invention has in addition to the features of the embodiment described immediately above:
A still further embodiment of the present invention has in addition to the features of the embodiment described immediately above, for each of said strain gauges, a Wheatstone bridge of four strain gauges, each mounted at a 45° angle with respect to the lengthwise axis of said block and each oriented at a 90° angle with respect to the adjacent strain gauges.
Another embodiment of the present invention has its first strain gauge mounting area formed as a recess extending inwardly from said front side toward said rear side, and its third strain gauge mounting area formed as a recess extending inwardly from said rear side toward said front side, said first and third recesses extending toward but not reaching each other, the bottoms of these recesses defining between them a first internal wall extending in a vertical plane in said lengthwise direction and having front and rear faces, one of said strain gauges mounted on each of said front and rear faces respectively of said first internal wall, and has its second and fourth strain gauge mounting areas formed as recesses, similarly as the recesses on said first and third strain gauge mounting areas, the bottom of these recesses defining between them a second internal wall similar to said first internal wall with strain gauges mounted on said front and rear faces respectively of said second internal wall, with said interface means adapted for receiving and averaging said output data from said first, second, third and fourth strain gauges and outputting a digital signal representative of said weight of said container and any load therein.
A still further embodiment of the present invention is a vehicle load weighing system using load cells as described above wherein said vehicle is a truck and said plurality of load cells comprises three load cells situated at the front, middle and rear on the right side respectively of said chassis and three additional load cells situated at the front, middle and rear on the left side respectively of said chassis.
The load cell and the vehicle load weighting system of the present invention have numerous elements which are combinable in numerous different configurations, many examples of these configurations being described in the specification and in the claims.
For example, the floating mount of this load cell is believed to be novel, wherein the load cell is fixed to one of the vehicle container and vehicle chassis components and is movably coupled to the other for relative movement thereto in the lengthwise, transverse and/or vertical direction. This may be achieved with slide or slip surfaces on the coupling means which are mounted to the vehicle chassis or to the vehicle container and engage force contact areas on the load cell block. With such a floating mount the load cell is subjected to far less stress and distortion.
Numerous combinations and variations of structure and function are possible from the disclosure herein, including some load cells with the floating mount and some without and some load cells with Wheatstone bridge strain gauge arrangements and some with other strain gauge configurations.
Another embodiment of the present invention is a method of enhancing environmental conservation through encouraging recycling through the use of variable pricing for waste pick-ups. General unsorted waste would be priced at a commercial rate. Segregated scrap, such as iron, copper, brass, paper and other such recyclable material would be picked up for free or at a greatly reduced rate. Through the incentive pricing, consumers and businesses would be encouraged to separate recyclable materials. Pricing policies could be implemented that provide rebates to customers who segregated commercially viable amounts of particularly valuable recyclable materials. Because quantities of unsorted garbage and recyclable materials can be measured accurately, with the on-board load weighing system, and records can be accurately kept on the Cab display controller, the method can be implemented without the need for costly and time consuming external scales.
The above and related objects, features and advantages of the present invention will be more fully understood by reference to the detailed description of the presently preferred, albeit illustrative, embodiments of the present invention when taken in conjunction with the accompanying drawings herein.
As used with trucks a typical load cell of the present invention is secured beneath a bottom surface of the truck container and upon a mounting surface of a beam of the truck chassis. With garbage trucks it is desirable to keep the truck container's center of gravity as low as possible to enhance stability and to render it easier for the truck operators to load and unload. Thus, instead of situating the load cell upon the top surface of the chassis beam, it is preferably mounted to a bracket at the side of the beam, but still positioned beneath the container, as will be described in further detail below.
The mounting arrangement of a typical load cell of the first embodiment is summarized as follows and described in detail thereafter.
As seen in
The load cell mounting scheme in this first embodiment is to secure each load cell 1006 to the truck container 1003, and then for each load cell via its slip surface 1017 (see
To allow this relative translational movement between the load cells and the truck chassis, each load cell has a slip contact surface which is in sliding engagement with a corresponding contact surface of a support bracket fixed to the chassis. The sliding contact arrangement of the load cell block 1006A and support bracket 1010A is seen best in
As seen in
As further seen in
As seen in
In each load cell the strain gauges are mounted in sets of four as seen in
The positioning of the strain gauges in the diagonal orientations, with four gauges forming a diamond pattern, and with a set of these gauges on both sides of the web results in a dual shear beam configuration which will produce a new level accuracy of strain measurement to reflect the weight added to the truck container. Accordingly each strain gauge is laterally spaced halfway between the upward and downward forces applied to the load cell, with cancellation of longitudinal, transverse and twisting forces that may be applied to the load cell block and thence to the strain gauges.
Each load cell block thus has two longitudinally spaced strain gauge units on internal webs with a set of strain gauges fixed on each side of the web, and potting substance to secure, seal and insulate the load cell unit. On the load cell block and slightly spaced from these strain gauges in a separate recess 1090 is located a microprocessor.
In the second load cell mounting embodiment illustrated in
The invention herein disclosed includes individual load cells, vehicle weighing systems which utilize the new load cells, and methods of determining the
incremental weight differences using the new load cells or using the new weighing system.
The electronic circuitry and characteristics of these load cells and characteristics of this system are described in detail below.
A key component of the system is the precision, intelligent load cell 1006. A precision intelligent load cell 1006, as shown in
When the strain gauge is used in a bridge configuration, it is possible to precisely detect the resistance changes. The signal level from the strain gauge 1205 is small, but low noise, high gain, highly stable operational amplifiers driving high resolution A/D converters, including averaging type of delta sigma A/D converters of 21 or higher bits of resolution yield high levels of accuracy and resolution, as described below. The strain gauges are oriented at a forty-five degree angle with respect to the principal axis of the load cell as shown in
The strain gauges are connected in a Wheatstone bridge configuration as shown in
This amplifier circuit uses differential input circuits 1210 to buffer the signals Vout+ and Vout− representing strain that originate in the strain gauge and also differential amplifiers 1212 to further process these signals to input them into A/D converter 1215. The amplifier circuits that buffer the strain gauge output should preferably have closely matched gain characteristics in order to ensure that the accuracy of the strain gauge signals, which are inputs to the differential inputs of the A/D converter 1215, are preserved. The A/D converter 1215 is preferably an averaging type of A/D converter, such as a delta sigma type A/D converter with a differential input such as the Linear Technology LTC 2413 A/D converter for the buffered strain gauge signals, although other types of A/D converter with 21 bits or better resolution may be used, with appropriate adjustments. It is desirable that the A/D converter contain internal digital filters to reduce noise on readings and auto-calibration circuits to improve A/D converter accuracy.
The preferred embodiment uses one A/D converter for each of the four strain gauge bridges in a load cell to avoid latency effects where an A/D converter will take several samples of the current strain gauge before it will reach full accuracy. If corresponding adjustments are made, the invention could be practiced with one A/D converter measuring the four strain gauges in one load cell. This would entail adding switches and making timing changes to the software. The details of these changes are well known to those skilled in the art.
The load cell microprocessor determines the temperature of the load cell by reading the value from a temperature sensor 1925, such as the Linear Technologies LT1722. This temperature reading is used by load cell software for temperature compensation of the weight readings as described below.
The load cell microprocessor 1905 communicates with the system cab controller 1008 through a controller area network (CAN) interface 1920. The CAN interface is a standard data and control interface that is used with many automotive and other systems and is fully described in the data sheet for Motorola Part No. MC68HC 08AZ60. The CAN interface provides a robust and stable interface which has excellent noise immunity, but the invention may be practiced with other system interfaces which currently exist or may be developed.
The load cell identifies its position within the truck weighing system through the use of a unique identifying resistor located in the chassis connector that connects module detect 1910 to ground. The load cell has an internal 10 kohm resistor in module detect 1910 that is connected to +Vac. These two resistors form a voltage divider which present a unique voltage based on the values of module detect 1910 and the identification resistor in the connector. An A/D converter that is located in microprocessor 1905 converts that unique analog identification voltage, which the microprocessor 1905 then compares to a look-up table under software control. The identification voltage thus indicates the position of the load cell which is used by both the load cell and the cabs display for fault identification and weight calculations. Generation of a list of resistor values suitable for location identification is based on resistor tolerances and the number of unique locations to be identified.
To achieve the high accuracy performance of the load cells, data from the four A/D converters in the load cell 1006 must be processed in order to filter noise and perform compensation for systematic errors as shown in
Furthermore, the sequence of the software steps may be changed, with appropriate adjustments, without departing from the spirit of the invention. Also, some steps may be omitted without departing from the scope of the invention.
All known prior truck load measuring systems have a single filter setting and the corresponding limitations or no filter. If the filter has a low filter value, the filter will have a fast response to changing loads but will not filter system noise very well. Using a high filter value results in very good noise and vibration immunity but at a loss of response speed to a changing load.
Filter Factor is a figure of merit and parameter for the adaptive filter, where the filter factor relates inversely to the effect that a new reading has on the output of the filter. A filter factor of one will result in the new reading passing through the filter unchanged. A filter factor of 64 will result in only 1/64th of the new data being reflected in the filter output.
If the current measured value is outside the window with respect to the current input value, the adaptive digital filter will decrease the filter factor 1413 in accordance with a set of rules until the minimum filter factor (usually one) is reached.
It should be noted that the rules by which the filter factor is increased or decreased need not be the same. Different rules can be used. The dynamics of the adaptive filter can be varied by changing the rules by which the filter factor is increased during stable readings (i.e., when the current instantaneous value is within the window) or decreased during rapidly changing readings (i.e., when the current instantaneous value is outside the window).
As examples of rules which may be used, the filter factor may be increased by increasing steps during periods of stable readings, so that maximum filtering will result. Also, the filter factor may be decreased by half for each current measurement outside the window, so that the filter factor is changed rapidly during periods of changing input value. Although this will reduce noise immunity, in will result in a faster settling time to the new input value.
The filter factor is used in calculating the next filtered value as follows:
As seen from the equation, the filter factor determines how much the current measurement changes the existing filtered value. If the filter factor equals one, the current filtered value is ignored and the next filtered value is simply the current measured value. The adaptive dynamic filter system of the present invention determines if a load is being added or removed and adjusts the filter factor value at a programmed rate to allow for fast response and then automatically increases the filter factor value when it detects that the load is stable. The method by which the invention determines a changing load and adjusts the filter value is more fully discussed below.
The curve labeled “Fast Filtered Weight” uses algorithms to reduce the filter factor on rapid changes of input weight and increase the filter factor for stable input weight conditions. The rules used to change the filter factor are those that are shown in
In the present invention, a programmable window is used. This window, as shown in
To improve noise immunity, it is also possible to compare the number of readings outside the window to a window counter in the filter 1407 through 1411, so that a number of readings outside the window must occur before the filter factor is changed in order to reduce the susceptibility of the filter to noise. One of ordinary skill in the art knows how to implement a window counter in the software that compares measurements to the window.
The divisor of 8 in the formula above is one possible increment rate limiter, and it is possible to adjust filter performance by changing the increment rate limiter. It should be evident to one of ordinary skill in the art that by making the increment rate limiter larger decreases the rate of change of the filter factor in response to out-of-window readings, while making the increment rate limiter smaller increases the rate of change to out-of-window readings. Reasonable experimentation would allow one of ordinary skill in the art to use other rules for changing the filter factor to tailor filter response to a particular set of requirements.
Assuming the current measurement was within the window, using the increment weight limiter value of 8 would result in the following series of filter factors: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 20, 22, 24, 27, 30, 33, 37, 40, 45, 50, 56, 64, 72, 81, 91, 102, 124, etc. The filter factor starts increasing by 2 after a filter factor of 16 is reached because the integer part of (fi−1)/8 is 2; by 3 after a filter factor of 24 because the integer part of (fi−1)/8 is 3; etc. A typical maximum filter factor value is 128, but larger values are possible with the appropriate adjustments. The filter factor, using the formula above, increases slowly at the start and then approaches its maximum allowable value reasonably fast. Without using the formula to increase the filter factor as successive measurements fell within the window, it would take 128 readings for the filter factor to reach 128, with the filter factor being increased by one with each new measurement.
To determine if the load is changing, the software checks the window counter to determine if the reading is outside of the window for the number of readings programmed into the window counter. If the requisite number of readings is outside the window, the filter factor is divided by two. Thus, assuming the filter factor starts at 128, and the readings remain outside the measurement window, the filter factor has values as follows: 128, 64, 32, 16, 8, 4, 2, 1—with the limitation that the minimum filter factor is one. A filter factor of one causes the new reading to be taken as the new value and thus no filtration or delay occurs.
Other increase and decrease routines can be developed to tailor the adaptive filter for faster or slower response and still be within the scope of the invention. The specific examples shown are for illustrating the preferred embodiment, but are not the only embodiments possible of the invention. For example, the invention could be practiced with a routine that decreases the filter factor during a load change with rules that are similar to those that are used when the load is stable but adjusted to reflect decrease of the filter factor.
This filter system will be slow to respond to load changes less than the window size. If the window is set for 2 pounds and a 12 ounce soda can is added, the system will slowly adjust to the added weight because the load cells do not detect an out-of-window condition to trigger a change in filter factor to speed the response. But if a 5 pound weight is added, more than the 2 pound window size, the system will sense a difference that is outside the window and decrease the filter factor in order to respond faster.
No external input is required for detection and response to changing conditions because the adaptive filter determines if a load is being placed on the truck by monitoring the weight measurements.
After the filtering of the raw analog-to-digital converter data, a temperature compensation is performed for the load cell zero voltage. This is performed using the following formula:
Z=z+a*T Δ +b*T Δ 2 +c*T Δ 3
Where “Z” is the resulting temp corrected Zero value; “z” is the basic Zero correction before temperature correction is added; “a” is the first order temperature correction factor; “b” is the second order temperature correction factor; “c” is the third order temperature correction factor, which may not be necessary; “TΔ” is the difference in temperature from the calibration temperature (i.e., the difference between the current temperature and the calibration temperature, which is a stored parameter representing the temperature of the load cell during the factory calibration); and “TΔ 2” is the temperature difference squared and “TΔ 3” is the temperature difference cubed. The values for a and b may be approximately −0.3 lb/deg and −0.007/deg/deg, respectively.
After the temperature correction of the load cell zero is performed, a temperature correction to the linearity characteristics of the load cell is performed using the following formula:
G=g+a*T Δ +b*T Δ 2 +c*T Δ 3
where “G” is the temperature corrected gain; “g” is the basic gain coefficient; “a”, “b”, and “c” are coefficients for the temperature correction that are determined during factory testing and “TΔ” is the temperature difference between the current temperature and the factory temperature when the temperature coefficients were originally determined.
Load cell weight=Z+G*S avg +C*R 2 +D*R 3
where: R=2 (Savg−½ calibration weight)/calibration weight; Savg is the average of the temperature compensated and filtered straining gauge measurements C is related to the error in reading at ½ full scale; D is a correction factor for third order effects, which has a different value for high weights and low weights.
Although the CAN bus is used in the preferred embodiment, any bus that allows multiple devices to be connected to a bus are within the scope of the invention. The use of a bus that allows more than one device to be connected to the controller through the bus simplifies the wiring required and reduces the amount of wire required. This improves the reliability and maintainability of the system. Furthermore, the use of module identification resistors, described above, enables the system controller to receive information from each module to locate the module's unique place on the truck.
As shown in
The generation of a command instruction set for implementing the various capabilities discussed herein on a CAN bus or a communications bus are well understood by those skilled in the art. It is necessary to list the parameters and data necessary to implement the functions and assign command instruction set codes to them. The protocol for command and data transfers is defined for most communications busses. The protocol for the CAN bus is published by Bosch.
The three major groups of commands and data that are needed are setting of operational parameters, and calibration parameters, and the reading of system configuration information. The operational parameters comprise commands to set up the analog-to-digital converter, set up the load cell filters or to read load cell temperature. Calibration parameters include the reading of load cell gain, zero, linearity correction and temperature correction. System configuration parameters include read back of load cell and tilt sensor data such as serial number, model number, date of manufacture and revision. Commands can also be assigned to read back load cell built-in-test status and position on the truck chassis.
The primary functions of the cab display are to provide system control and to perform the system weight measurements under control of the operator. As a system controller, the cab display reports faults within the system based on diagnostics run in the load cells and in the other units. The cab display also reports faults in the CAN bus that impair ability to communicate with the other units.
As shown in
After the load is on the truck the operator presses the “Pickup Done” button. At this point, the cab display software captures the final system state, consisting of the readings from each of the load cells and the tilt sensor. The cab display software then calculates a pickup weight based on the difference between the start-of-pickup state, and the end-of-pickup state, and truck tilt angles. Load cell software compensates for temperature on the weights that are transmitted to the cab display.
The system will display the weight added during the “Pickup,” with the appropriate resolution. The ranges for the system can be for example: 0 to 1,000 pounds; 0 to 10,000 pounds; and 0 to 50,000 pounds. Resolutions for these ranges would be 1 pound, 2 pounds and 10 pounds, respectively. The range for each pick-up is based on the load that is picked up. Since this truck weighing system uses load cells that provide high resolution and accuracy for measured weights, it is possible to provide different ranges for different load pickups while using the same basic internal load cell resolution. As the pickup load goes over the upper limit for a range, the cab display simply reformats the display using the resolution appropriate for the new range and displays the correct weight. Since the load cells have sufficient precision for each range, no change to load cell operation is needed.
The load weight that is measured is an incremental weight that is based on the system state (i.e., the readings from each of the load cells and the tilt sensor before and after the load). Since these two readings, before and after the load is added, are made within a short time of each other, most systemic errors due to temperature, drift, creep, or truck chassis twisting due to an uneven surface are not a significant error in the calculation of load weight. Temperature zero effects are eliminated due to temperature compensation and the close proximity in time of the two measurements that establish the pickup load. Gain temperature effects are significant in the invention. Since the truck weighing system makes incremental measurements based on the difference of load cell readings before and after a load is picked up, long term zero temperature effects are not a significant issue, but short term effects (i.e., temperature change during the 5 to 30 minutes for picking up a load) may become significant. Long term gain temperature effects are significant and care is taken in designing the strain gauge, amplifiers and A/D converter circuits to maintain a gain shift of less than or equal to 0.001% over a 5 degree Celsius temperature change.
The system will not allow for display of a pickup weight that is beyond the maximum limits of the scale system. Thus a 50,000 pound range can not fully accommodate a 50,000 pound pick up if the truck already has 10,000 pounds on it. The same applies to the 10,000 pound range. If the truck has 45,000 pounds loaded on the system, then only 5,000 pounds can be added in the 10,000 pound range. At no point is the system going to allow any range to extend beyond the system capacity. The actual weight on the truck is only used as a limit for making sure that operation outside of the scale range does not occur, and for driver information that the truck is reaching or has reached the legal payload limits. The scale capacity will normally be higher than the legal truck payload rating. Thus the condition of going beyond the scale capacity should not occur in normal operation.
Upon completion of the Pickup, the system will log the pickup weight and other required information. Other information will be current weight on the truck along with time and date and customer ID code and related information. A ticket will be printed if the customer desires a printed Pickup ticket. Pickup weight will be printed using the resolution for the range.
The load cells are required to have an internal resolution and stability of the area of 0.1 pound or better. The system is required to display the weight to 1 pound resolution for the smaller load ranges/intervals, and is required to do this at any weight that is on the truck to meet the 1 pound count for the 1000 pound range.
Since a truck may be parked on a sloping surface before starting each pickup, tilt sensors are provided for tilt correction. The worst case condition is a pickup where the truck has a tire on a curb and this causes a shift in the weight distribution of the container on the load cells because the truck chassis is not in a fixed plane. The result is 50% or more weight shifting between load cells, as the truck chassis is twisted. This is not a problem with the present system since the load cells have sufficient extra range to still make accurate measurements.
Other options that the operator can choose at the cab display include selection of a transaction printout so that a printed receipt can be provided to the customer.
Because of the accuracy of the truck weighing system described, total truck weight can be calculated from the sum of each of the individual pickup weights. This provides a periodic check of accuracy because the truck will be weighed on a daily basis as the loads are brought to the dump. The truck weight kept by the on-board system can be compared with the external scale at this time. Because of the superior accuracy and precision, errors in weight on each load are reduced and billings for pickups can be more accurate.
The system design of this on-board truck weighing system provides other advantages. Because of the open architecture of the system, future expansion is possible to add, for example, global positioning system (GPS) data to the pickup record, to provide an additional documentation of the pickups. The system can also be connected to computers in the factory to diagnose failures within the system. The connection can be made using wired connectors to the CAN interface or wireless adapters that are connected to the CAN interface and interface with wireless adapters in the shop or work site.
Now that the preferred embodiments of the present invention have been shown and described in detail, various modifications and improvements will become readily apparent to those skilled in the art. Accordingly, the spirit and scope of the present invention is to be construed broadly and limited only by the appended claims and not the foregoing specification.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4364279 *||Dec 31, 1980||Dec 21, 1982||Allegany Technology, Inc.||Shear beam load cell system|
|US4516646 *||Mar 7, 1983||May 14, 1985||Aktiebolaget Bofors||Shear beam load cell with built-in bearing elements|
|US4666003 *||Sep 17, 1985||May 19, 1987||Stress-Tek, Inc.||On-board load cell|
|US4744254 *||Dec 12, 1986||May 17, 1988||Pfister Gmbh||Force measuring device|
|US4958526 *||Oct 10, 1989||Sep 25, 1990||Flintab Ab||Force measuring device with zero adjustment|
|US5294756 *||May 5, 1992||Mar 15, 1994||Endress+Hauser Inc.||Conveyor belt scale apparatus|
|US5319161 *||Dec 24, 1992||Jun 7, 1994||Pitney Bowes Inc.||Mechanism for preventing overload on a weighing scale|
|US5440078 *||Jul 19, 1993||Aug 8, 1995||Schrran Engineering, Inc.||Portable scale with load cell|
|US5811738 *||Nov 8, 1996||Sep 22, 1998||Larry D. Santi||Trunnion-mounted weight measurement apparatus|
|US6005199 *||Aug 28, 1998||Dec 21, 1999||Kawatetsu Advantech Co., Ltd.||Load cell and load detecting apparatus|
|US6295878 *||Dec 13, 1999||Oct 2, 2001||Bertec Corporation||Particular strain gage orientation for a six component load measurement device|
|US6331682 *||Jan 14, 2000||Dec 18, 2001||Hottinger Baldwin Measurements, Inc.||Tank weigh module with excess motion restraint|
|US6583367 *||May 18, 2001||Jun 24, 2003||Trw Inc.||Parallelogram load sensing apparatus for a seat belt webbing|
|US6766685 *||Jan 22, 2003||Jul 27, 2004||Timothy B. Foley||Method of sensing and correcting brake functions in moving heavy truck brake systems using instrumented anchor pins|
|US6898989 *||Dec 1, 2000||May 31, 2005||Abb Ab||Load cell|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7644628||Jun 4, 2009||Jan 12, 2010||Loadstar Sensors, Inc.||Resistive force sensing device and method with an advanced communication interface|
|US7705251 *||Jun 28, 2007||Apr 27, 2010||Herrmann Ag||Rest element for a lifting platform|
|US8354602 *||Jan 21, 2010||Jan 15, 2013||Halliburton Energy Services, Inc.||Method and system for weighting material storage units based on current output from one or more load sensors|
|US8452525 *||Sep 22, 2011||May 28, 2013||Eurocopter, Deutschland GmbH||Weighing system and methods of operating such weighing system|
|US8581457 *||May 25, 2012||Nov 12, 2013||Ntn Corporation||Wheel support bearing assembly with sensor and in-wheel motor integration|
|US8841566 *||Apr 15, 2011||Sep 23, 2014||Stress-Tek, Inc.||Load cell assemblies for measuring off-center loads|
|US20110174551 *||Jul 21, 2011||Lucas Bruce C||Weighing Assembly for Material Storage Units|
|US20110253462 *||Oct 20, 2011||Keith Reichow||Load cell assemblies for off-center loads and associated methods of use and manufacture|
|US20120116660 *||Sep 22, 2011||May 10, 2012||Eurocopter Deutschland Gmbh||Weighing system and methods of operating such weighing system|
|US20120229004 *||Sep 13, 2012||Ntn Corporation||Wheel support bearing assembly with sensor and in-wheel motor integration|
|US20140262557 *||Mar 15, 2013||Sep 18, 2014||Illinois Tool Works Inc.||Load cell|
|U.S. Classification||177/211, 177/229, 73/862.627|
|International Classification||G01G19/12, G01G3/13|
|Cooperative Classification||Y10S177/09, G01G19/12, G01G23/3735|
|European Classification||G01G23/37W2, G01G19/12|
|Feb 3, 2009||AS||Assignment|
Owner name: THE AIS GROUP, LLC,CALIFORNIA
Free format text: SECURITY AGREEMENT;ASSIGNOR:DYNAMIC DATUM, LLC;REEL/FRAME:022191/0212
Effective date: 20021231
|Jan 7, 2013||REMI||Maintenance fee reminder mailed|
|May 16, 2013||FPAY||Fee payment|
Year of fee payment: 4
|May 16, 2013||SULP||Surcharge for late payment|